# Music-and-Culture-Technology-Lab/omnizart

Omniscient Mozart, being able to transcribe everything in the music, including vocal, drum, chord, beat, instruments, and more.

Repository: https://github.com/Music-and-Culture-Technology-Lab/omnizart
Canonical: https://ross.abutalabs.com/products/omnizart
Homepage: https://music-and-culture-technology-lab.github.io/omnizart-doc/
Language: Python
License: MIT
License Family: permissive
Topics: music-transcription, drum-transcription, vocal, beat-tracking, chord, music-information-retrieval
Last push: 2026-05-31T07:09:13+00:00

## Health v2 (maintenance only)
Score: 88/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 85, release rhythm 86, longevity 100
- inputs: {"age_days": 2198, "days_push": 94, "days_rel": 94, "gap_med": 0, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1964, forks 135 (observed 2026-08-28T04:05:59.760259+00:00)

## What it is
Omnizart is a Python library and CLI for automatic music transcription, transcribing pitched instruments, vocal melody, chords, drum events, and beat from polyphonic audio. It covers the full deep learning lifecycle including dataset downloading, feature preprocessing, model training, transcription, and sonification, with pre-trained checkpoints provided.

## Use cases
- transcribe piano notes from a song to midi
- extract drum events from audio
- detect chord progression in a track
- transcribe vocal melody from a song
- track beats in music audio
- train a custom music transcription model

## When to choose
- you need automatic music transcription of instruments, vocals, chords, drums, or beats
- you want pretrained checkpoints for immediate transcription without training
- you want to train or evaluate your own transcription models in Python

## When to avoid
- you need real-time low-latency transcription
- you need a GUI-based music editor rather than a library/CLI
- you work outside Python and don't want to use Docker

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, audio-processing, cli, deep-learning
- domain: machine-learning, audio, deep-learning
- platform: python, cli, cross-platform
- tags: music-transcription, music-information-retrieval, drum-transcription, vocal-transcription, chord-recognition, beat-tracking, pretrained-models, mir, audio, docker

## Member repositories
- Music-and-Culture-Technology-Lab/omnizart (main) score 88

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:59.760259+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:05:29.774852+00:00, confidence not recorded.
  - readme: https://github.com/Music-and-Culture-Technology-Lab/omnizart (fetched 2026-08-28T04:05:59.760259+00:00, sha bb5f2aa42386)
  - homepage: https://music-and-culture-technology-lab.github.io/omnizart-doc/ (fetched 2026-08-29T10:45:09.266222+00:00, sha f44f8e434465)
  - registry_pypi: https://pypi.org/pypi/omnizart/json (fetched 2026-08-29T10:45:09.275622+00:00, sha 85d4d291ea82)
- Data as of 2026-08-30T08:39:29.467469+00:00.